APPLICANT'S ABSTRACT: Current views of cocaine abuse that focus on the incentive properties of drugs and their associated stimuli hold promise for guiding research that may lead to effective treatment of cocaine abuse. Impressive evidence implicating a particular neural substrate in mediating the incentive properties of psychomotor stimulants and other drugs of abuse, i.e., the mesolimbic dopamine system, potentiates the likelihood of swift and important contributions of neurobiological research toward this goal. In this proposal, using the high temporal and spatial resolution of single-cell electrophysiology to elucidate neural mechanisms of cocaine self-administration, questions are formulated within the incentive motivational framework. Recordings will be obtained from both the origin and termination of the mesolimbic system, i.e., from dopamine neurons in the ventral tegmental area (VTA) and from their target neurons in the nucleus accumbens septi (NAS). Rats in which lever pressing is maintained on a fixed ratio 1 (FRl) schedule of reinforcement by intravenous infusion of cocaine (0.7 mg/kg) exhibit regular inter-infusion intervals (mean = 6.7 min). Each reinforced lever press (RLP) is followed by a prolonged period of focused stereotypy that gradually yields to an increasing percentage of time spent in locomotion. Eventually, one or more approaches toward the lever, characterized as drug-seeking behaviors, culminate in another RLP. Single-unit recordings from mesolimbic target neurons in the NAS show cyclic patterns of firing synchronized to this behavioral cycle. While certain firing patterns conform to interpretations based on pharmacokinetics and/or on relationships of firing to motor behavior, other specific patterns do not. For example, in many cases transitions from maximum to minimum firing rate (or vise versa) are complete within < 2 sec of the RLP, in advance of any significant infusion-related increase in cocaine levels in mesolimbic areas (explicitly addressed in Specific Aim 2 by introducing on selected trials a brief delay between the RLP and the infusion). These rapid transitions in firing could reflect the cessation of drug-seeking behavior upon execution of the RLP. Supporting this interpretation are results of a preliminary experiment (the basis of Specific Aim 1), using a multiple schedule, in which a sharp elevation in firing immediately preceded the RLP during FR1 responding. Next, when the temporal pattern of cocaine infusions from the FR1 phase was presented non-contingently, infusions were no longer immediately preceded by 1) approach to and depression of the lever or 2) the sharply elevated firing; both returned when the FR1 schedule was reinstated. Thus, certain phasic NAS firing patterns do not appear to reflect only pharmacokinetics but may be related to drug- seeking behaviors, consistent with the hypothesis that mesolimbic system neurons process information regarding conditioned incentive stimuli. In Specific Aim 3, to test this hypothesis more directly than in the above studies, in which the processing of such stimuli and the timing of infusions were controlled by the animal, a programmed auditory tone will serve as a discriminative stimulus (SD) signalling the end of a variable time-out period following each RLP. After responding has been brought under stimulus control, SD-evoked alterations in firing will be assessed. In light of the considerable explanatory power of incentive- motivational theory as a framework for approaching a neurobehavioral understanding of drug abuse, the present design offers a unique opportunity to test hypotheses directly on neurons focal to this approach, i.e., the mesolimbic dopamine system.